FG-SMOTE: Fuzzy-based Gaussian synthetic minority oversampling with deep belief networks classifier for skewed class distribution.
Putta HemalathaGeetha Mary AmalanathanPublished in: Int. J. Intell. Comput. Cybern. (2021)
Keyphrases
- class distribution
- minority class
- majority class
- class imbalance
- training data
- highly skewed
- imbalanced datasets
- imbalanced data
- cost sensitive
- training set
- training samples
- imbalanced data sets
- cost sensitive learning
- test set
- misclassification costs
- training examples
- class labels
- restricted boltzmann machine
- highly imbalanced
- unlabeled data
- maximum likelihood
- test data
- concept drift
- labeled data
- multi class
- base classifiers
- dimensionality reduction
- support vector machine
- feature space
- support vector